ISSN: 2522-266X
Research Article · Volume 5, Issue 1 (2026) · pp. 1–10 Open Access
Islamic economics is an economic system where individuals conduct their economic activities within the framework of Islamic law, prioritizing moral, ethical, and social dimensions. Islamic finance is built upon fundamental principles such as the prohibition of interest, risk sharing, and the distinction between halal and haram, and has shown rapid growth and diversification on a global scale, especially in the last quarter century. Islamic stock markets consist of the shares of companies operating in accordance with Sharia principles and are referred to as the “Participation Index” in Türkiye. The increasing number of studies on participation indices in the literature reveals that Islamic finance is an increasingly important research area. This study aims to predict the price movements of the BIST Participation All Shares Index (XKTUM) using machine learning methods. In the analysis, the USD/TRY exchange rate, the price of gold per ounce, the Brent oil price, and the overnight TL interest rate were used as independent variables to predict the XKTUM index for the period 16.11.2021–02.12.2025. LightGBM and XGBoost algorithms were used to evaluate the forecasting performance. The dataset contains 1,015 daily observations, and the last 20% of the sample (10.02.2025–02.12.2025) is used as the test set. LightGBM produced a lower forecasting error than XGBoost, with SMAPE values of 17.47% and 20.48%, respectively. The USD/TRY exchange rate is the dominant predictor in both models. Both models tracked the index in the early part of the test period, while they failed to follow the strong upward trend after mid-2025 because tree-based models cannot extrapolate beyond the range of the training data. A six-month forecast based on ARIMA projections of the explanatory variables points to limited variation in XKTUM and is presented as an illustrative scenario.
Participation IndexMachine LearningPrice PredictionLightGBMXGBoost
© 2026 The Author(s). Published by the Eurasian Review of Business and Finance. This is an open access article distributed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).